Sodium Vitamin C Cotransporter SVCT2 is Expressed In Hypothalamic Glial Cells

Glia

Sodium vitamin C cotransporter SVCT2 is expressed in hypothalamic glial cells María De Los Angeles García 1, Katherine Salazar 1, Carola Millán 1, Federico Rodríguez 1, Hernán Montecinos 1, Teresa Caprile 1, Carmen Silva 1, Christian Cortes 1, Karin Reinicke 1, Juan Carlos Vera 4, Luis G. Aguayo 2, Juan Olate 3, Benedicto Molina 5, Francisco Nualart 1 * 1Departamento de Biología Celular, Facultad de Ciencias Biológicas, Universidad de Concepción, Concepción, Chile 2Departamento de Fisiología, Facultad de Ciencias Biológicas, Universidad de Concepción, Concepción, Chile 3Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias Biológicas, Universidad de Concepción, Concepción, Chile 4Departamento de Fisiopatología, Facultad de Ciencias Biológicas, Universidad de Concepción, Concepción, Chile 5Departamento de Ciencias Básicas, Facultad de Medicina, Universidad de la Frontera, Temuco, Chile email: Francisco Nualart ([email protected]) *Correspondence to Francisco Nualart, Laboratorio de Neurobiología Celular, Departamento de Biología Celular, Facultad de Ciencias Biológicas, Universidad de Concepción, Casilla 160C, Concepción, Chile Funded by: FONDECYT, Chile; Grant Number: 1010843, 3020007 Concepcion University, Chile; Grant Number: DIUC 201.035.002-1.0, DIUC 203.031.094-1 La Frontera University, Chile; Grant Number: DIUFRO-EP2120 Keywords SVCT2 • vitamin C • glia • tanycytes • hypothalamus • ependymal cells • pars tuberalis Abstract Kinetic analysis of vitamin C uptake demonstrated that different specialized cells take up ascorbic acid through sodium-vitamin C cotransporters. Recently, two different isoforms of sodium-vitamin C cotransporters (SVCT1/SLC23A1 and SVCT2/SLC23A2) have been cloned. SVCT2 was detected mainly in choroidal plexus cells and neurons; however, there is no evidence of SVCT2 expression in glial and endothelial cells of the brain. Certain brain locations, including the hippocampus and hypothalamus, consistently show higher ascorbic acid values compared with other structures within the central nervous system. However, molecular and kinetic analysis addressing the expression of SVCT transporters in cells isolated from these specific areas of the brain had not been done. The hypothalamic glial cells, or tanycytes, are specialized ependymal cells that bridge the cerebrospinal fluid with different neurons of the region. Our hypothesis postulates that SVCT2 is expressed selectively in tanycytes, where it is involved in the uptake of the reduced form of vitamin C (ascorbic acid), thereby concentrating this vitamin in the hypothalamic area. In situ hybridization and optic and ultrastructural immunocytochemistry showed that the transporter SVCT2 is highly expressed in the apical membranes of mouse hypothalamic tanycytes. A newly developed primary culture of mouse hypothalamic tanycytes was used to confirm the expression and function of the SVCT2 isoform in these cells. The results demonstrate that tanycytes express a high-affinity transporter for vitamin C. Thus, the vitamin C uptake mechanisms present in the hypothalamic glial cells may perform a neuroprotective role concentrating vitamin C in this specific area of the brain. © 2004 Wiley-Liss, Inc.

-------------------------------------------------------------------------------Received: 1 December 2003; Accepted: 8 July 2004 Digital Object Identifier (DOI) 10.1002/glia.20133 About DOI

INTRODUCTION In the nervous system, vitamin C has different functions; however, among these is the role as an antioxidant defending against free radicals (Wilson, [1987]; Halliwell, [1992]; Makar et al., [1994]; Siow et al., [1998]; Witenberg et al., [1999]; Brahma et al., [2000]; Rice, [2000]). Recently, vitamin C has also been shown to be involved in modulation of nitric oxide neurotransmission in the hypothalamic area (Karanth et al., [2000]). Kinetic analysis of vitamin C uptake has demonstrated that specialized cells, such as neurons, melanocytes, astrocytes, chromaffin cells, and fibroblasts, take up ascorbic acid (AA), the reduced form of vitamin C (Spector and Lorenzo, [1973]; Siliprandi et al., [1979]; Diliberto et al., [1983]; Wilson, [1989]; Welch et al., [1993]; Spielholz et al., [1997]; Zreik et al., [1999]; Malo and Wilson, [2000]; Castro et al., [2001]). Two different isoforms of sodium-vitamin C cotransporters (SVCT1 and SVCT2) have been cloned (Faaland et al., [1998]; Daruwala et al., [1999]; Rajan et al., [1999]; Tsukaguchi et al., [1999]; Wang et al., [1999], [2000]), these genes appear to belong to the family of solute carriers 23A (SLC23A1 and SLC23A2). Both SVCT proteins mediate high-affinity Na+-dependent L-AA transport and are necessary for the uptake of vitamin C in many tissues. SVCT1 is a 604-amino acid protein that is expressed in epithelial cells of kidney, intestine, and liver (Faaland et al., [1998]; Daruwala et al., [1999]; Tsukaguchi et al., [1999]). SVCT2 is a 592-amino acid protein that shares 65% homology to SVCT1 and has been detected mainly in brain and eyes (Rajan et al., [1999]; Tsukaguchi et al., [1999]). The expression of SVCT2 has been demonstrated in cultures of astrocytes but not in brain tissue (Berger and Hediger, [2000]). Recently, a short SVCT2 isoform has been discovered that does not function as transporter but rather acts as a dominant-negative inhibitor of AA transport through protein-protein interaction (Lutsenko et al., [2004]). Thus, vitamin C transporter expression and function in brain cells may be finely regulated.

correspond to types and 1. The 1 tanycytes are located in the lateral lower area of the ventricle and develop elongated cell processes that form a bow through the arcuate nucleus and reach the lateral sulcus of the infundibular region in the lateral hypothalamus. In this area, the endfeet contact luteinizing hormone-releasing hormone (LH-RH) terminals, which are involved in hormone release to the hypophyseal portal vessels for delivery of LH-RH to the pituitary gland. We previously demonstrated that mouse and 1 tanycytes express GLUT1 at high levels, with the highest levels of GLUT1 present in the cell processes making contact with the local capillary walls (García et al., [2001], [2003]). Similar results have been shown by Peruzzo et al. ([2000]) in rat hypothalamus. Because and 1 tanycytes do not appear to have a clear barrier function and are located in a region of the hypothalamus very low in astrocytes, we postulated that these cells may have astrocyte-like functions and facilitate metabolic coupling between glia and neurons and recycling of the vitamin C in the hypothalamic area (García et al., [2001], [2003]). High concentrations of AA are found in the fetal and adult brain (Spector and Lorenzo, [1973]; Chatterjee et al., [1975]; Schenko et al., [1982]; Oke et al., [1987]; Rice, [2000]; Hediger, [2002]). Certain brain locations, including the hippocampus and hypothalamus, consistently show higher AA values compared with other structures within the central nervous system (Oke et al., [1987]). However, molecular and kinetic analyses addressing the expression of SVCT transporters in cells isolated from these specific areas of the brain have not been performed.

It has been postulated that vitamin C should be recycled within the brain in order to maintain high concentrations in glial cells and neurons (Hediger, [2002]). In the recycling mechanism, glial cells (mainly astrocytes in brain cortex and hippocampus) take up the oxidized form of vitamin C, dehydroascorbic acid through GLUT1. Upon transport into astrocytes, dehydroascorbic acid is reduced to AA. AA may also be released to the extracellular fluid by astrocytes stimulated under physiological conditions (Wilson Vitamin C is highly concentrated in the hypothalamic area (Oke et et al., [2000]). Thus, SVCT2 and GLUT1 transporters are al., [1987]; Karanth et al., [2000]); however, data on the extremely important in vitamin C homeostasis in brain cells expression and function of SVCT2 in neuronal or glial cells in the (Wilson, [1987]; Qutob et al., [1998]; Korkok et al., [2000]). hypothalamic area are not available. It has been postulated that neurons and glial cells may have different mechanisms for the In the present study, we demonstrate the expression of SVCT2 uptake and recycling of vitamin C (Rice, [2000]). Furthermore, transporter in hypothalamic glial cells (tanycytes) by in situ most cells take up the oxidized form of vitamin C, hybridization and immunofluorescence associated with confocal dehydroascorbic acid (DHA), through the facilitative glucose microscopy. Ultrastructural immunohistochemistry confirmed that transporters 1 (GLUT1, gene name SLC2A1), a mechanism that SVCT2 is localized in the cellular membranes of the apical may be central in vitamin C recycling in the brain (Rose, [1988]; microvilli and blebs of the cells. Furthermore, the SVCT2 Vera et al., [1993]; Welch et al., [1995]; Agus et al., [1997], function was confirmed by kinetic assays in cultured tanycytes. [1999]; Rumsey et al., [1997]; Nualart et al., [2003]). We observed that AA uptake in tanycytes is inhibited with ouabain and phloretin. Additionally, SVCT2 expression in most Tanycytes are specialized hypothalamic glial cells localized in neurons was confirmed by in situ hybridization. Finally, our study circumventricular organs such as the median eminence (Flament- confirms that SVCT2 is not expressed in astrocytes or endothelial Durand and Brion, [1985]; García et al., [2001], [2003]). They can cells of the blood-brain barrier (BBB). be classified into at least four types: 1, 2, 1, and 2 (Akmayev and Fidelina, [1974]). The 2 tanycytes form tight junctions and MATERIALS AND METHODS participate in the formation of the cerebrospinal fluid (CSF)median eminence barrier (Chauvet et al., [1995]). However, these Immunocytochemistry and Confocal Microscopy cells express glucose transporter GLUT1, a metabolic barrier Mice (C57BL/J6) brains were dissected and fixed immediately by marker for cells that have tight junctions (Peruzzo et al., [2000]; immersion in Bouin's solution. Fixation in situ was performed by García et al., [2001], [2003]), at a very low concentration. The vascular perfusion (Nualart et al., [1991]). Samples were tanycytes located in the lower walls of the third ventricle dehydrated in graded alcohol solutions and embedded in paraffin.

Frontal sections (5 m) of the hypothalamic area were mounted on poly-L-lysine-coated glass slides. For immunohistochemical analysis, we used an affinity-purified goat polyclonal antibody raised against a peptide mapping near the amino terminus of SVCT2 of rat origin (Santa Cruz Biotechnology, Santa Cruz, CA). Additionally, we used an affinity-purified rabbit polyclonal antibody raised against a C-terminal sequence of the glucose transporter GLUT1 (Alpha Diagnostic, San Antonio, TX). Sections were incubated overnight at room temperature in a humid chamber, with anti-SVCT2 (1:100) or anti-GLUT1 antibodies (1:100) diluted in a Tris-HCl buffer (pH 7.8) containing 8.4 mM sodium phosphate, 3.5 mM potassium phosphate, 120 mM NaCl, and 1% bovine serum albumin (BSA). Additionally, in some immunohistochemical experiments, an antibody anti-GFAP was used (1:200 (DAKO, Carpinteria, CA). After extensive washing extensively, the sections were incubated for 2 h with Cy2-conjugated affinity-purified donkey anti-goat IgG (1:200; Jackson ImmunoResearch, West Grove, PA) at room temperature. Alternatively, anti-goat IgG (1:50; DAKO) labeled with peroxidase (1:30; DAKO) was used as a secondary antibody. For confocal laser microscopy analysis, the tissue sections were incubated with propidium iodide in the absence of RNase for cellular staining. As negative controls for SVCT1 and SVCT2, we used both primary antibodies pre-absorbed with the respective peptides used to elicit them and with preimmune serum.

HCl (pH 7.5), 1 mM ethylenediaminetetraacetic acid (EDTA), 1× Denhardt's solution, 10% polyethylene glycol 8000, 10 mM DLdithiothreitol (DTT), 500 g yeast tRNA/ml, 50 g/ml heparin, 500 g/ml DNA carrier, and 1:20 to 1:100 dilutions of riboprobe) were added to each slide. The slides were covered with glass coverslips and placed in a humidified chamber at 42°C overnight. After removal of the coverslip, the slides were rinsed in 4× salinesodium citrate (SSC) and washed twice for 30 min at 42°C. The slides were washed at 37°C for 30 min each in 2× SSC, 1× SSC, and 0.3× SSC. Visualization of digoxigenin was performed by incubation with a monoclonal antibody coupled to alkaline phosphatase (anti-digoxigenin alkaline phosphatase Fab fragments diluted 1:500; Boehringer-Mannheim) at room temperature for 2 h. Nitroblue tetrazolium (NBT) chloride and 5-bromo-4-chloro-3indolyl-phosphate (Boehringer Mannheim) were used as substrates for the alkaline phosphatase. Controls included use of the sense riboprobe and omission of the probe. Western Blot Mouse total brain extracts, total brain cell membranes, and total hypothalamus cell membranes were obtained by homogenizing the cells in 0.3 mM sucrose, 3 mM DTT, 1 mM EDTA, 100 g/ml phenylmethylsulfonyl fluoride (PMSF), 1 g/ml pepstatin A, and 2 g/ml aprotinin. Total membranes were collected by high-speed centrifugation. 90 g of membrane protein was loaded in each lane, separated by polyacrylamide gel electrophoresis (PAGE), using a 5-15% polyacrylamide gradient in the presence of sodium dodecyl sulfate (SDS), transferred to nitrocellulose membranes (García et al., [2003]), and probed with the affinity-purified anti-SVCT2 antibody (1:200; antiserum raised against a peptide mapping near the N-terminus of rat/human SVCT2). The antibody was diluted in TBS (10 mM Tris-HCl, pH 7.4, and 149 mM sodium chloride) containing 5% nonfat milk, 0.05% Tween-20 (TBS-blot buffer) and incubated with the membranes overnight at room temperature. Membranes were incubated in a rabbit anti-goat IgG antibody (1:500, DAKO) for 2h and in goat anti-rabbit IgG coupled to peroxidase (1:10,000, Pierce Biotechnology, Rockford, IL) for 2 h in TBS-blot buffer. The reaction was developed with enhanced chemiluminiscence using the enhanced chemiluminescence (ECL) Western blotting analysis system (Pierce Biotechnology).

Ultrastructural Immunohistochemistry Brain tissues were immersed for 2 h in fixative containing 2% paraformaldehyde, 0.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4. The samples were dehydrated in dimethylformamide and embedded in London Resin Gold (Electron Microscopy Science, Washington, DC). Ultrathin sections were mounted on uncoated nickel grids and processed for immunocytochemistry (Peruzzo et al., [2000]). For immunostaining, the anti-SVCT2 antibody (1:100) was diluted in incubation buffer Tris-HCl (pH 7.8) containing 8.4 mM sodium phosphate, 3.5 mM potassium phosphate, 120 mM NaCl and 1% BSA. After extensive washing, the ultrathin sections were incubated for 2 h at room temperature with anti-rabbit IgG 10-nm colloidal gold labeled (1:20). Uranyl acetate/lead citrate was used as contrast and samples were analyzed using a Hitachi H-700 electron microscope with 125-200-kV accelerating voltage. Deglycosylation Proteins from hypothalamic membrane extracts (98 g) were diluted in 100 mM sodium phosphate buffer, pH 7.4, containing In Situ Hybridization A cDNA of 0.6 kb subcloned in pCR-4-Blunt-TOPO and 50 mM EDTA, 0.1% SDS, and 1% -mercaptoethanol. After 5 min encoding the rat SVCT2 was used to generate sense and antisense at 95°C, 0.6 U N-glycosidase F (Boehringer-Mannheim) and 1% digoxigenin-labeled riboprobes. RNA probes were labeled with Triton X-100 were added to the samples. The reaction mixture digoxigenin-UTP by in vitro transcription with SP6 or T7 RNA was then incubated at 37°C, for 4 h. Control samples were polymerase following the manufacturer's procedure (Boehringer- prepared by omitting the enzyme from the incubation mixture. Mannheim, Mannheim, Germany). In situ hybridization was The samples were separated by PAGE, using a 5-15% performed on mouse hypothalamic frontal sections mounted on polyacrylamide gradient in the presence of SDS, transferred to poly-L-lysine-coated glass slides. The sections were baked at nitrocellulose membranes (García et al., [2003]), and probed with 60°C for 1 h, deparaffinized in xylene, and rehydrated in graded the affinity-purified antibody anti-SVCT2. Membranes were ethanol. Following proteinase K treatment (5 min at 37°C in incubated with a rabbit anti-goat IgG antibody (1:500, DAKO) for phospate-buffered saline [PBS], 1 g/ml), the tissue sections were 2 h and in goat anti-rabbit IgG coupled to peroxidase (1:10,000, fixed with 4% paraformaldehyde at 4°C for 5 min, washed in cold Pierce Biotechnology) for 2 h in TBS-blot buffer. The reaction PBS, and then acetylated with 0.1 M triethanolamine-HCl (pH was developed with DAB and H2O2 (Nualart and Rodríguez, 8.0) and 0.25% acetic anhydride at room temperature for 10 min. [1996]). We used this protocol to detect the positive reaction and After a brief wash, the sections were incubated in pre- not ECL method, because the lower intensity observed with DAB hybridization solution at 37°C for 30 min, and then 25 l of and H2O2 allowed us to estimate more accurately the SVCT2 hybridization mix (50% formamide, 0.6 M NaCl, 10 mM Tris- molecular mass change after deglycosylation.

Primary Cultures and Immunocytochemistry Tanycytes were obtained from C57BL/J6 mice. The forebrains were removed from 19-day-old embryos. Hypothalamic areas were digested for 10 min with 0.25% trypsin in phosphate buffer 0.1 M (pH 7.4) and homogenized with a fire-polished Pasteur pipet. The cells were plated at 300,000 cells/ml onto six-well plates with or without glass coverslips. Cells were incubated for about 1 month and fed every 3 days with minimal essential medium (MEM) containing 10% fetal bovine serum (FBS), 4 mM L-glutamine, and 100 U/ml penicillim and 100 g/ml streptomycin. These cultures were characterized using several antibodies (García et al., [2003]). For immunocytochemistry, the tanycytes were fixed with 4% paraformaldehyde in PBS. The cells were incubated overnight with anti-SVCT2 or anti-SVCT1 (1:100 Santa Cruz Biotechnology). Cells were then incubated with goat anti-rabbit IgG-FITC. Reverse Transcription-Polymerase Chain Reaction The poly(A) RNA from cultured tanycytes and hypothalamic area tissues was isolated using the Oligotex direct kit (Qiagen, Valencia, CA). For reverse transcription-polymerase chain reaction (RT-PCR), 0.5- 1 g of RNA was incubated in 20 l reaction volume containing 10 mM Tris pH 8.3, 50 mM KCl, 5 mM MgCl2, 20 U RNase inhibitor, 1 mM dNTPs, 2.5 M of random hexanucleotides, and 50 U of MuLV reverse transcriptase (Perkin-Elmer, Branchburg, NJ) for 10 min at 23°C followed by 30 min at 42°C and 5 min at 94°C. For amplification, a cDNA aliquot in a volume of 12 l containing 20 mM Tris, pH 8.4, 50 mM KCl, 1.6 mM MgCl2, 0.4 mM dNTPs, 0. 04 U of Taq DNA polymerase (Gibco, Rockville, MD), and 0.4 M primers was incubated at 94°C for 4 min, 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s for 35 cycles. The following primers (based on human sequence NM-001101.2) were used to analyze the expression of -actin: forward primer 5-GCT CGT CGT CGA CAA CGG CTC-3 and reverse primer 5-CAA ACA TGA TCT GGG TCA TCT TCT C-3 (expected product 360 bp). The following primers (based on human sequence AF164142, gene HGNC ID:10973) were used to analyze the expression of SVCT2 transporter, forward primer 5-TTC TGT GTG GGA ATC ACT AC-3 and reverse primer 5-ACC AGA GAG GCC AAT TAG GG-3 (expected product 339 bp). The following primers (based on human sequence AF170911, gene HGNC ID:10974) were used to analyze the expression of SVCT1 transporter, forward primer 5-GCC CCT GAA CAC CTC TCA TA-3 and reverse primer 5ATG GCC AGC ATG ATA GGA AA-3 (expected product 360 bp). Vitamin C Uptake Analysis Tanycyte cultures were carefully selected under the microscope to ensure that only plates showing uniformly growing cells were used at 200,000 cells/well. The cells were incubated in buffer 15 mM HEPES (N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid), 135 mM NaCl, 5 mM KCl, 1.8 mM CaCl2, 0.8 mM MgCl2 at room temperature for 30 min. Uptake assays were performed in 500 l of incubation buffer containing 0.1-0.4 Ci of 1-14C-L-AA (specific activity 8.2 mCi/mmol), to a final concentration of 5-300 M. The Michaelis constant, Km, was calculated using the Lineweaver-Burk analysis. The data represent means ± SD of three experiments with each determination done in duplicate. In inhibition experiments, statistical comparison between two or more groups of data was carried out using analysis of variance

(ANOVA, followed by Bonferroni post-test). P < 0.05 was considered statistically significant. RESULTS In Situ Hybridization Analysis of SVCT2 Transporter in Brain Cells and Glial Hypothalamic Cells We analyzed the expression of SVCT2 at the mRNA level by in situ hybridization using digoxigenin-labeled cRNA probes specific for SVCT2 (Fig. 1A-H). An intense hybridization signal was observed in neurons of the mouse brain. The most intense signal was detected in neurons of the brain cortex, cerebellum, hippocampus, hypothalamus, and habenular nucleus (Fig. 1A,C-F, and data not shown). A clear hybridization was observed in the ependymal cells of the upper zone of the third ventricle (Fig. 1A, small arrows); however, the ependymal cells of the lateral ventricle were almost negative (Fig. 1B, arrow labeled E). Similarly, there was no labeling in endothelial cells of the BBB and astrocytes present in different brain regions (Fig. 1A,B). To confirm the presence of negatively labeled astrocytes in the tissue sections, we performed in situ hybridization for SVCT2 in combination with immunohistochemical analysis, using antibodies against GFAP an astrocyte marker (Fig. 1C,D). Additionally, to confirm the presence of negatively-labeled endothelial cells and astrocytes in the tissue sections we performed in situ hybridization for SVCT2 in combination with immunohistochemical analysis using antibodies against GLUT1 (endothelial cells marker) and GFAP (Fig. 1E). The choroid plexus cells showed an intense positive reaction; however, some cells were almost negative (Fig. 1B, arrow labeled CPC) indicating a differential expression of SVCT2 in choroid plexus cells. In the hypothalamic area, the glial cells that cover the ventricular walls, and tanycytes, showed an intense hybridization with SVCT2 antisense (Fig. 1F). In the median eminence staining was observed in 2 tanycytes (Fig. 1G,H), which are cells that form tight junctions and participate in the formation of CSF-median eminence barrier (Fig. 1J, arrows). There was no labeling in astrocytes present in the median eminence, which were identified by an anti-GFAP antibody (Fig. 1H, arrows). This situation contrasted with the intense level of labeling observed in the pars tuberalis cells (Fig. 1G,H). Control experiments using a sense riboprobe in serial sections revealed no labeling of the tanycytes, confirming the specificity of the reaction with SVCT2 antisense riboprobe (Fig. 1I).

SVCT2 mRNA in these cells (Fig. 2D). In situ hybridization for SVCT2 in combination with immunohistochemical analysis using antibodies against GFAP demonstrated that the hybridization signal for SVCT2 co-localized with the GFAP staining in the apical cytoplasm of the tanycytes, but not in GFAP-positive processes (Fig. 2E, arrows and 2I, white arrows). Additionally, the astrocytes localized in the subventricular area were negative for SVCT2 antisense (Fig. 2F,G, white arrows). No immunoreaction for GFAP was detected in astrocyte-like cells of the lower lateral wall of the third ventricle where the arcuate nucleus neurons and and 1 tanycytes are located (Fig. 2E). These results suggest that astrocytes are totally replaced by tanycytes in this hypothalamic region. In the hypothalamus, our hybridization data indicate that the ventricular glial cells and the neurons possess the highest expression of SVCT2 transporter mRNA; however, the presence of a given mRNA in certain tissues does not always correspond to expression of the protein. Thus, we performed an immunohistochemical analysis using specific antibodies for SVCT2.

Figure 1. SVCT2 mRNA detection by in situ hybridization. Frontal section of mouse brain probed with a SVCT2 antisense riboprobe. A: A high hybridization signal was observed in neurons of the habenular nucleus. B: The choroidal plexus cells of the lateral ventricle show intense staining. C,D: In situ hybridization for SVCT2 (C) in combination with immunohistochemical analysis by incubation with an antibody against GFAP (astrocyte marker) (D). E: Hippocampal area. In situ hybridization for SVCT2 in combination with immunohistochemical analysis by simultaneous incubation with antibodies against GFAP (astrocyte marker) and GLUT1 (endothelial cells marker). The hybridization signal was only detected in neurons. F-H: The and 1 tanycytes present positive hybridization with the antisense SVCT2 riboprobe (F). In and 1 tanycytes the positive reaction is observed in the proximal area of the cells (F, arrows). Also the 2 tanycytes show a positive reaction for SVCT2 riboprobe (G). H: In situ hybridization for SVCT2 in combination with immunohistochemical analysis using GFAP antibody. The astrocytes (arrows) of the median eminence present positive immunoreaction for GFAP (arrows); however, they are negative for SVCT2 hybridization. I. The sense riboprobe for SVCT2 was used as a control. J: 2 tanycytes observed by using scanning electron microscopy. The lateral membranes of the cells are connected by tight junctions (arrows). III V, third ventricle. A, astrocytes; AN, arcuate nucleus; CPC, choroid plexus cells; DMN, dorsomedial nucleus; E, ependymal cells; HN, habenular nucleus; ME, median eminence; PT, pars tuberalis; V, blood vessel; VMN, ventromedial nucleus. Scale bars = 100 m in A,B,D-F; 500 m in C; 15 m in D.

In the lateral hypothalamus, the ciliated ependymal cells were almost negative when probed with SVCT2 antisense (Fig. 2A,D,F, arrow labeled E); however, in the transitional area of the ventricular wall the negative ciliated ependymal cells were interchanged with positive non-ciliated tanycytes (Fig. 2B,D,H, arrows labeled ). Similarly, the 1 tanycytes that have blebs differentiation in the apical membranes of the cells (Fig. 2C) showed an intense hybridization signal (Fig. 2D, arrows labeled 1). There were no apparent differences in the labeling intensity of and 1 tanycytes, suggesting the presence of equivalent levels of

Figure 2. In situ hybridization for SVCT2 in combination with immunohistochemical analysis using GFAP antibody in hypothalamic area. A-C: Scanning electron microscopy of the ventricular wall associated with the hypothalamic basal area. D: Frontal section of mouse hypothalamus probed with an antisense SVCT2 riboprobe. A strong hybridization signal was observed in neuroendocrine neurons and tanycytes (high magnification in H); however, the ependymal cells were negative (high magnification in F). E: In situ hybridization for SVCT2 and immunohistochemical analysis using a GFAP antibody. The processes (arrows) of and 1 tanycytes present positive immunoreaction for GFAP (high magnification in I, white arrows). The astrocytes of the subventricular area are also positive (high magnification in G, white arrows). E, ependymal cells; Tn, tanycytes. Scale bars = 15 m in A-C; 25 m in D,E; 12 m in F-I.

SVCT2 Transporter Is Preferentially Expressed in Hypothalamic Glial Cells The specificity of the antibody was confirmed by using Western blot analysis. When we studied the affinity of an antiserum raised against a peptide mapping near the amino terminus of rat/human SVCT2 a single band of 75 kDa was detected in mouse total brain protein extract (Fig. 3, lane A). SVCT2 detection resulted in a

marked increase in the width and intensity of the anti-SVCT2 immunoreaction band migrating at 75 kDa in total brain and hypothalamic samples prepared from cellular membrane extracts (Fig. 3, lanes B and C, respectively). To compare the molecular size of SVCT2 with another solute transporter expressed in brain cells, we used a specific antibody raised against the glucose transporter GLUT1. A single band of 55 kDa was detected in mouse total brain cellular membrane extracts that corresponds to the expected size for this glucose transporter (Fig. 3, lane G). Using our electrophoretic condition (5-15% polyacrylamide concentration gradient gels) we always observed a sharp band for SVCT2 and GLUT1, two glycosylated transporters. This result suggests that the electrophoretic conditions concentrate the glycosylated proteins in one specific area of the gel. As negative controls for SVCT2 and GLUT1, we utilized both primary antibodies pre-absorbed with the respective peptides used to elicit them (Fig. 3, lanes D and H, respectively). To confirm that the protein recognized by anti-SVCT2 correspond to the glycosylated SVCT2 transporter we incubated the hypothalamic protein extract with N-glycosidase F, an enzyme that cleaves all types of asparagine bound N-glycans from the proteins (Plummer et al., [1984]; Nualart and Rodríguez, [1996]). After incubation with Nglycosidase F the band of 75 kDa could no longer be detected, instead, a band of 66 kDa was observed (Fig. 3, lanes E and F). Our results confirm that SVCT2 is glycosylated in vivo.

Figure 3. Comparative Western blot analysis of SVCT2 and GLUT1 in brain and hypothalamic region. A-F: Western blot analysis using an affinity-purified goat polyclonal antibody raised against a peptide mapping near the amino terminus of SVCT2 of rat origin. G,H: Western blot analysis using an affinity-purified rabbit polyclonal antibody raised against a C-terminal sequence of the glucose transporter GLUT1. A: Protein extracted from mouse total brain. B,G,H: Protein extracted from mouse total brain membranes. C-F: Protein extracted from total hypothalamus membranes. D: control. F: total hypothalamus membranes treated with N-glycosidase F enzyme that cleaves all types of asparagine bound N-glycans from the proteins. The samples were separated by polyacrylamide gel electrophoresis using a 5-15% polyacrylamide gradient in the presence of sodium dodecyl sulfate and transferred to nitrocellulose membranes. SVCT2 transporter showed an apparent molecular mass of 75 kDa (A-C,F). SVCT2 showed a molecular mass of 66 kDa after deglycosylation (F). Anti-GLUT1 showed a single band of 55 kDa.

Initially, we analyzed the expression of SVCT2 in the hypothalamus of adult rat by immunohistochemistry using antiSVCT2 antiserum and a secondary antibody conjugated with peroxidase (Fig. 4A,B). Additionally, immunofluorescence and confocal microscopy were applied in mouse brain (Fig. 4C,D). Immunohistochemical analysis led us to demonstrate a prominent anti-SVCT2 immunoreactivity in the rat median eminence and the hypothalamic ventricular area (Fig. 4A,B). The axons of the

hypothalamo-neurohypophyseal tract were strongly immunostained (Fig. 4A). A similar immunoreaction was observed in neurons of the paraventricular nucleus (Fig. 4A, inset). This reaction was not observed in other neurons of the hypothalamus. The tanycytes of the ventricular area were also positives (Fig. 4B, arrows). No immunoreactivity for SVCT2 transporter was observed in endothelial or other glial cells of the hypothalamus (Fig. 4A,B). The different cells of the pars tuberalis presented a low immunostaining (data not shown). We carefully examined the immunoreaction in the tanycytes in order to identify precisely the cell types that express SVCT2. The and 1 tanycytes showed positive anti-SVCT2 immunoreactivity, with the 1 cells showing the strongest immunoreaction (Fig. 4B, arrows). 2 tanycytes that are involved in the formation of the median eminence-CSF barrier were negative for SVCT2 (Fig. 4A, small arrows). Identical data were also derived by immunofluorescence analysis of mouse hypothalamus. The cells presented in the hypothalamic area and median eminence were identified using propidium iodide staining (Fig. 4C,D, red color), and SVCT2 was detected using a secondary antibody labeled with Cy2 (Fig. 4C,D, green color). The immunoreactive material was associated with neuroendocrine axons of the median eminence (Fig. 4A). The 1 tanycytes presented a clear immunoreaction with anti-SVCT2 at the level of the proximal part of the cells (Fig. 4D, short arrows). To compare SVCT2 and GLUT1 expression in 1 tanycyte we performed an immunofluorescence analysis in mouse serial hypothalamic sections. As we have determined previously, and 1 tanycytes showed positive anti-GLUT1 immunoreactivity, with the 1 cells showing the highest immunostaining (García et al., [2001], [2003]). The 1 tanycytes presented a marked immunoreaction with anti-GLUT1 at the level of the cell processes which form a bow through the arcuate nucleus (Fig. 4F, short arrows), which are negative with anti-SVCT2 (Fig. 4D). SVCT2 and GLUT1 immunoreaction was totally negative when the antibodies were pre-incubated with the relevant peptides (Fig. 4E, and data not shown).

Figure 4. Immunohistochemical analysis of SVCT2 expression in rat and mouse hypothalamus. A,B: Frontal sections of rat hypothalamus immunostained with anti-SVCT2 (1:50) using a secondary antibody conjugated with peroxidase. C,D: Frontal sections of mouse hypothalamus immunostained with anti-SVCT2 (1:50) using a secondary antibody labeled with Cy2. E,F: Frontal sections of mouse hypothalamus immunostained with anti-GLUT1 (1:100) using a secondary antibody labeled with Cy2. SVCT2 immunoreactivity is detected in the proximal part of 1 tanycytes (A,B,D, short arrows) and in neuroendocrine axons to the median eminence (A,C). A similar immunoreaction is observed in some neuroendocrine neurons of the paraventricular nucleus (A, inset). SVCT2 immunoreactivity is observed in the proximal part and short processes of the rat 1 tanycytes (B, arrows). The same result is obtained in mouse hypothalamus (D, arrows). In these hypothalamic serial sections, the whole cells were stained with propidium iodide (C,D, red) and SVCT2 was detected using immunofluorecence analysis (C,D, green). A SVCT2 signal is detected in neuroendocrine axons of the median eminence (C) and in the proximal part of 1 tanycytes (D, short arrows). The 2 tanycytes are negative for SVCT2 (C, short arrows). A similar immunofluorescence analysis was performed in hypothalamic serial sections using anti-GLUT1 (E,F). GLUT1 is detected mainly localized in cell processes (F, arrows) and endfeet of 1 tanycytes (F). The proximal area of the 1 tanycytes is also positive (F). III V, third ventricle; ME, median eminence. Scale bars = 300 m in A; 25 m in B-F.

Ultrastructural immunohistochemistry confirmed that the SVCT2 transporter is located in the apical membrane of the tanycytes (Fig. 5A-F). The highest immunoreaction was observed in 1 tanycytes that had the gold particles concentrated in the microvilli and blebs of the cells (Fig. 5B,D). The and 2 tanycytes were almost negative for SVCT2 confirming the data obtained with immunofluorescence and confocal microscopy (Fig. 5E,F). A labeling pattern was not detected in the cytoplasmic area of the tanycytes, neuronal membranes or endothelial cells (Fig. 5B,D, and data not shown).

Figure 5. Ultrastructural immunocytochemistry of hypothalamic ventricular wall. A-F: Frontal sections through the mouse medial basal hypothalamus. Immunohistochemical analysis using anti-SVCT2 antibody and anti-IgG labeled with 10-nm gold particles. A,B: Transitional area between and tanycytes. B,C: Dorsal wall of the infundibular recess showing the 1 tanycytes. E,F: Ventral wall of the infundibular recess showing the 1 tanycytes. The immunoreaction is mainly observed in the apical membranes of the 1 tanycytes. The gold particles are concentrated in the membranes of the microvilli and blebs (B,D, arrows). The neuronal membranes and the tanycytes located in the ventral wall of the infundibular recess are almost negative (B,D,F). III V, third ventricle; IR, infundibular recess. Scale bars = 1 m in A,C,E; 0.5 m in B,D,F.

SVCT2 Is Highly Expressed in Primary Tanycyte Cultures We seeded our cultures with cells obtained by thoroughly dissecting the prenatal mouse hypothalamic area. After five weeks in culture without passage, we selected flasks containing confluent cell growth with an elongated appearance (Fig. 6A). Most cells had a polarized morphology that consisted of a wide proximal cytoplasmic region containing the nucleus and a long basal process (Fig. 6B). Immunohistochemical analysis revealed an intense positive reaction with anti-vimentin and anti-p75NTR

in cultured tanycytes. Anti-GFAP produced a negative immunoreaction (García et al., [2003]). The cellular characteristics of cultured tanycytes were totally different from those of ependymal cells in culture which present an epithelial aspect in conditions of confluent growth (Fig. 6C). The isolated cells were short in appearance and with cilia present in the apical area (Fig. 6D, arrow).

Specific antibodies for AA transporters revealed expression of SVCT2 in cultured tanycytes (Fig. 6E). The anti-SVCT2 immunoreactivity was intense and was evenly distributed throughout the cell population (Fig. 6E). SVCT2 was detected in cellular membranes and was specially concentrated in the cytoplasmic area around the nucleus; however, the cell processes were not stained with anti-SVCT2 (Fig. 6E). To control for the specificity of the anti-SVCT2 immunoreactivity, we used primary antibodies pre-absorbed with the blocking peptides which eliminated the reaction (Fig. 6F). The expression of sodium-vitamin C cotransporter isoforms (SVCT1 and SVCT2) in cultured tanycytes and hypothalamic tissue was analyzed by RT-PCR with primers specific for each form. The conditions were optimized using RNA from human brain tissue as a control for the expression of SVCT2 and RNA from human intestine for SVCT1. As determined by gel electrophoresis, the amplified DNA band corresponded to 339 bp, the expected sizes for the amplification product of human SVCT2 (Fig. 6G, lane 2). For SVCT1 transporter the expected size for the amplification product was 360 bp (Fig. 6G, lane 9). No amplification product was observed in samples in which the cDNA synthesis step was performed in the absence of reverse transcriptase, indicating the absence of DNA contamination in the RNA preparation (Fig. 6G, lane 6). Using mRNA isolated from mouse adult hypothalamus, 1-day-old hypothalamus and cultured tanycytes, we amplified the cDNA of the SVCT2 isoform. As judged by their migration, the amplified DNA bands corresponded to approximately 339 bp, the expected size for the amplification product of mouse SVCT2 (Fig. 6G, lanes 3, 4, and 5, respectively). SVCT1 was not detected in samples isolated from human brain, cultured mouse tanycytes and hypothalamus (Fig. 6G, lanes 7 and 8, and data not shown, respectively). The 360-bp band was only detected in control samples of human intestine (Fig. 6G, lane 9). The expression and concentration of -antin indicates the high quality of the RNAs used for performing the analysis (Fig. 6G). Our results indicate the expression of SVCT2 transporters in hypothalamus, cultured tanycytes and human brain tissue.

Figure 6. Immunohistochemical analysis and reverse transcription-polymerase chain reaction (RT-PCR) of SVCT2 in cultured tanycytes. The cells were obtained from mouse hypothalamus at 19 days of gestation. A-D: Tanycytes and ependymal cells after 5 weeks in culture. The tanycytes are organized in monolayers and show an elongated form (A). B: High magnification of a single cell using Nomarski optics. The cell shows a polarized aspect with an apical expanded area (arrow) and a single process (asterisk). C,D: Ependymal cells in culture. D: High magnification of a single cell using Nomarski optics. The cell shows a narrow apical area (arrow). E: Tanycytes inmunostained with anti-SVCT2 and a secondary Cy2labeled antibody. F, Negative control for SVCT2. The antibody specific for SVCT2 was pre-absorbed with the peptide used to induce the antiserum. G: RTPCR for SVCT2 mRNA using mRNA isolated from cultured tanycytes. Lane 1, DNA 100 base pair standard; lane 2, amplified sequence of 339 bp using mRNA from adult human brain and primers for SVCT2 cDNA; lane 3, RT-PCR product obtained using mRNA isolated from mouse adult hypothalamus; lane 4, RT-PCR product obtained using mRNA isolated from 1 day-old mouse hypothalamus; lane 5, RT-PCR product obtained using mRNA isolated from cultured tanycytes (sequence analysis confirmed SVCT2 expression); lane 6, negative control; lane 7, RT-PCR products obtained using mRNA from adult human brain and primers for SVCT1 cDNA; lane 8, RT-PCR product obtained using mRNA from tanycytes in culture; lane 9, RT-PCR product (360 bp) obtained using mRNA from human intestine. Scale bars = 20 m in A,C; 2 m in B,D; 6 m in E,F.

Kinetic Characterization of the Ascorbic Acid Transport in Cultured Tanycytes As there were no previous studies addressing the kinetic properties of SVCT2 transporter in tanycytes, we examined AA transport in cultured tanycytes. The transport of AA was linear with time for the first 5 min (Fig. 7A) and reached a plateau at about 20 min (data not shown). The initial velocity of AA uptake was 50 pmol/106 cells × min. The total uptake in 5 min was 255 pmol × 106 cells (Fig. 7A). To test the sodium dependence of AA transport the NaCl in the incubation buffer was replaced with choline chloride. Transport of AA by tanycytes was decreased by at least 90% in the absence of sodium, showing a initial velocity of 5 pmol/106 cells × min and the uptake reached a plateau of 45 pmol/106 cells in 5 min (Fig. 7A). To characterize the transport of AA in tanycytes further, we studied the effect of temperature (Fig. 7B). Transport was almost basal at 4°C. However, when the cells were incubated at 22°C, the velocity of the AA transport was 20 pmol/106 cells × min. The AA transport was almost three to four times higher when the uptake experiment was performed at 37°C (74 pmol/106 cells × min). These results strongly suggested

cotransport of sodium with AA by tanycytes. Analysis of the sodium ion effect on AA transport at 50 M AA revealed that this transporter was strongly activated by sodium (Fig. 7C). The sodium ion effect was of a cooperative nature as indicated by the slightly sigmoidal shape of the concentration / velocity curve. This conclusion was corroborated by the result of a Hill plot of the data. This plot yielded a straight line with a slope (Hill coefficient) of 1.74 (Fig. 7D). Next we examined the characteristics of AA transport in tanycytes based on the initial evidence of its sodium dependence. A detailed concentrationresponse study showed that the transport of AA approached saturation near 60 M substrate (Fig. 8A). Analysis of the transport data by the Lineweaver-Burk method revealed the presence of one functional component, involved in AA transport by tanycytes (Fig. 8B). The transporter showed an apparent transport Km value of 20 M and a Vmax value of 40 pmol/min per 106 cells. Competition and inhibition experiments revealed that 10 mM deoxyglucose, 10 mM L-glucose, 10 M cytochalasin B or E failed to affect AA transport, indicating that the AA transporters expressed by tanycytes are functionally unrelated to the dehydroascorbic acid transporters. However, ouabain at a concentration of 20 M inhibited the transport by more that 80% (Fig. 9) and 100 M phloretin inhibited the transport by 30% (data not shown).

Figure 9. Acute effect of inhibitors on the initial rates of ascorbic acid uptake. Tanycytes were pretreated with the different inhibitors for 5 min. The ascorbic acid (50 M) uptake was evaluated at 5 min of incorporation at 37°C. The absolute value corresponding to 100% is 310 ± 25 pmol × 106 cells. The results were obtained with 135 mM NaCl at 37°C. Data represent means ± SD of two experiments. The asterisks indicate that values are significantly different from control (*P < 0.01 and **P < 0.01).

DISCUSSION

Figure 7. Analysis of vitamin C transport in primary cultures of mouse tanycytes. A: Time course of the uptake of 50 M ascorbic acid in the presence of NaCl (open circles) or replacing NaCl with choline chloride (solid circle). B: Uptake at 5 min of 50 M ascorbic acid between 4 and 37°C. The plot represents the initial velocity of ascorbic acid uptake in each temperature point analyzed. C: Velocity of uptake of ascorbic acid (100 M) as a function of the concentration of extracellular sodium at 37°C. D: Hill plot for the effect of sodium in the uptake of ascorbic acid. Data represent the mean ± SD of three experiments.

Figure 8. Kinetic analysis of the ascorbic acid transports in primary cultures of mouse tanycytes. A: Concentration-response curve for the transport of ascorbic acid at 37°C. B: Lineweaver-Burk plot for the substrate dependence of ascorbic acid transport.

Several studies have postulated that SVCT2 is principally expressed in neurons and cells of the choroid plexus (Tsukaguchi et al., [1999]; Berger and Hediger, [2000]; Yan et al., [2001]). By functional and RT-PCR analysis of the AA uptake performed in neurons we demonstrated by that SVCT2 is expressed in cultured neurons (Castro et al., [2001]). A Km value of 8 M for the uptake of vitamin C by neurons was also observed, suggesting the expression of a high-affinity transporter for AA in these cells (Castro et al., [2001]). A similar high-affinity transporter has been detected in human melanoma, fibroblast and CaCo-2 cells (Welch et al., [1993]; Spielholz et al., [1997]; Maulen et al., [2003]). In this study, the SVCT2 expression in most neurons was confirmed using an in situ hybridization technique. Glial cells were mostly negative for SVCT2; however, we detected a clear hybridization signal in ependymal cells from the third ventricle and in glial cells of the hypothalamic zone, the tanycytes. To determine the high specificity of our riboprobe for SVCT2, we observed the intense staining present in choroid plexus cells. These cells showed a clear hybridization signal in choroid plexus of the lateral, third, and fourth ventricles, demonstrating that SVCT2 may be involved in the uptake of vitamin C by choroid plexus cells. Additionally, our study confirms that SVCT2 is not expressed in astrocytes or endothelial cells of the BBB. Furthermore, we performed a detailed immunohistochemical and in situ hybridization analysis of the SVCT2 expression in tanycytes. Tanycytes are specialized cells localized in the hypothalamic ventricular walls. Herein, we precisely localized and identified a subpopulation of tanycytes that express high levels of SVCT2 transporter in the hypothalamus of rat and mouse brain. Our localization data were confirmed by functional analysis. From our immunohistochemical data, we are able to extrapolate that tanycytes express SVCT2 in vivo. RT-PCR and immunoblot analyses have demonstrated the expression of

SVCT2 transporter in brain hypothalamus and in cultured tanycytes. The kinetic analysis and the inhibition data consistently indicate the expression of SVCT2 transporter in primary cultures of tanycytes. Thus, we established that the elevated expression of SVCT2 is restricted to and 1 tanycytes, cells that are located in the lower lateral wall of the third ventricle and do not participate in the formation of the brain-cerebrospinal barrier. It is possible that 2 tanycytes may express a different vitamin C transporter isoform that could replace SVCT2 in these cells. Our data, however, are consistent with the very low-to-nonexistent expression of SVCT1 transporter. In conclusion, our data indicate that SVCT2 is the main AA transporter expressed by ependymal tanycytes and therefore plays a central role in providing these cells with vitamin C, this is crucial, given that the brain is not involved in vitamin C biosynthesis, since the enzymes that participate in the synthesis of this vitamin are not expressed in brain (Chatterjee et al., [1975]).

Associated with the expression of SVCT2 in tanycytes, we have demonstrated that the expression of SVCT2 in and tanycytes is associated with a high expression of GLUT1 (García et al., [2001]), a classical transporter of glucose and of the oxidized form of vitamin C, dehydroascorbic acid (Vera et al., [1993]; Maulen et al., [2003]; Nualart et al., [2003]). Interestingly, the tanycytes concentrate GLUT1 in the endfeet of the tanycytes that contact the LH-RH terminals (García et al., [2001], [2003]). Considering that tanycytes are not involved in transcellular transport of glucose (García et al., [2001]), the overexpression of GLUT1 may be associated with dehydroascorbic acid uptake and vitamin C recycling (Nualart et al., [2003]). The AA present in the extracellular space of the hypothalamus should be oxidized to dehydroascorbic acid after interaction with NO (regulation of LHRH) or after antioxidative defense. High concentrations of dehydroascorbic acid in brain tissue might induce brain toxicity (Rice, [2000]). In response to this, it is believed that brain cells may take up the oxidized form of vitamin C using the glucose SVCT2 showed a preferential localization in microvilli and blebs transporter GLUT1 and reduce the molecule to AA in order to of 1 tanycytes, which strongly suggest that these cells obtain AA minimize its deleterious effects (Patterson and Mastin, [1951]; mainly from the CSF. It has been demonstrated that the CSF has a Rose et al., [1992]; Siushansian et al., [1997]; Rice, [2000]). high concentration of AA (500 M; 10 times more in peripheral blood). The lack of SVCT2 in 2 tanycytes strongly suggests that Overall, the tanycytes concentrate vitamin C intracelullarly by the these cells do not have the capacity to transport reduced vitamin C expression of the SVCT2 transporter and the uptake of the from the CSF or from the blood vessel of the median eminence. reduced form of vitamin C, AA. Additionally, the tanycytes may concentrate vitamin C via the high expression of the glucose Our data raise the question of the role of SVCT2 in the transporter GLUT1, that is involved in the uptake of the oxidized physiology of and 1 tanycytes. The immunohistochemical data form of vitamin C, dehydroascorbic acid (Vera et al., [1993]; demonstrated that 1 tanycytes have high levels of SVCT2 Nualart et al., [2003]). transporter, which endow these cells with the capacity to take up reduced vitamin C very efficiently. Although no information is Acknowledgements currently available concerning the antioxidant properties or intracellular function of vitamin C in tanycytes, ultrastructural This work is dedicated to Professor Esteban Rodríguez. analysis has shown the presence of large mitochondria (Akmayev Universidad Austral de Chile, Chile. The authors thank Ximena and Popov, [1977]; Rodríguez et al., [1979]), suggesting that these Koch, Universidad de Concepción, for her technical support and cells may have an elevated metabolic activity and oxidant Dr. Bruno Peruzzo, Universidad Austral de Chile, for his support production. Thus, the vitamin C may be concentrated inside the in ultrastructural immunohistochemistry. tanycytes using the SVCT2 transporter preferentially located in the microvilli of the cells in order to protect the tanycytes from References oxidative damage. Additionally, the hypothalamic tanycytes may have astrocytes-like function. The astrocytes have high Agus DB, Gambhir SS, Pardridge WM, Spielholz C, Baselga J, intracellular concentration of vitamin C inside the cells that may Vera JC, Golde DW. 1997. Vitamin C crosses the blood-brain be released in specific areas of the brain tissue after specific barrier in the oxidized form through the glucose transporters. J neurotrophic or neurotransmitter stimulation (Wilson et al., Clin Invest 100: 2842-2848. Links [2000]; Korkok et al., [2000]). Associated with this function, it Agus DB, Vera JC, Golde DW. 1999. Stromal cell oxidation: a has been demonstrated that AA has a high extracellular mechanism by which tumors obtain vitamin C. Cancer Res 59: concentration in neuroendocrine areas of the hypothalamus 4555-4558. Links (Schreiber and Trojan, [1991]; Das et al., [1993]). Thus, it has Akmayev IG, Fidelina OV. 1974. Morphological aspects of the been suggested that AA might control LH-RH release in the hypothalamic-hypophyseal system. V. The tanycytes: their lateral area of the hypothalamus, possibly by its antioxidant relation to the hypophyseal adrenocorticotrophic function. An properties (Karanth et al., [2000]). LH-RH release is stimulated enzyme-histochemical study. Cell Tissue Res 152: 403-410. Links by nitric oxide (NO) produced from NOergic interneurons that Akmayev IG, Popov AP. 1977. Morphological aspects of the release the soluble gas in juxtaposition to the LH-RH terminal hypothalamic-hypophysial system. VII. The tanycytes: their (Rettori et al., [1993]; Bhat et al., [1995]). Associated with these relation to the hypophyseal adrenocorticotrophic function. An terminals are the endfeet of tanycytes that separate the LH-RH ultrastructural study. Cell Tissue Res 180: 263-282. Links axons from the hypophyseal portal vessels (Bergland and Page, Berger UV, Hediger MA. 2000. The vitamin C transporter [1979]; Hökfelt et al., [1988]). AA released from the tanycyte SVCT2 is expressed by astrocytes in culture but not in situ. endfeet may have inhibitory functions blocking NOergic NeuroReport 11: 1395-1399. Links stimulation of LH-RH release by chemically reducing the NO Bergland RM, Page RB. 1979. Pituitary-brain vascular relations: a released by the cells. new paradigm. Science 204: 18-24. Links

Bhat GK, Mahesh VB, Lamar CA, Ping L, Aguan K, Brann DW. 1995. Histochemical localization of nitric oxide neurons in the hypothalamus: association with gonadotropin-releasing hormone neurons and co-localization with N-methyl-D-aspartate receptors. Neuroendocrinology 62: 187-197. Links Brahma B, Forman R E, Stewart EE, Nicholson C, Rice ME. 2000. Ascorbate inhibits edema in brain slides. J Neurochem 74: 1263-1270. Links Castro M, Caprile T, Astuya A, Millán C, Reinicke K, Vera JC, Vásquez O, Aguayo L, Nualart F. 2001. High-affinity sodiumvitamin C co-transporters (SVCT) expression in embryonic neurons. J Neurochem 78: 815-823. Links Chatterjee IB, Majumder AK, Nandi BK, Subramanian N. 1975. Synthesis and some major functions of vitamin C in animals. Ann NY Acad Sci 258: 24-47. Links Chauvet N, Parmentier ML, Alonso G. 1995. Transected axons of adult hypothalamo-neurohypophysial neurons regenerate along tanycytic processes. J Neurosci Res 41: 129-144. Links Daruwala R, Song J, Koh WS, Rumsey SC, Levine M. 1999. Cloning and functional characterization of the human sodiumdependent vitamin C transporters hSVCT1 and hSVCT2. FEBS Lett 460: 480-484. Links Das PC, Das KP, Bagchi K, Dey CD. 1993. Evaluation of tissue ascorbic acid status in different hormonal states of female rat. Life Sci 52: 1493-1498. Links Diliberto EJ Jr, Heckman GD, Daniels AJ. 1983. Characterization of ascorbic acid transport by adrenomedullary chromaffin cells: evidence for Na+-dependent co-transport. J Biol Chem 258: 12886-12894. Links Faaland CA, Race JE, Ricken G, Warner FJ, Williams WJ, Holtzman EJ. 1998. Molecular characterization of two novel transporters from human and mouse kidney and from LLC-PK1 cells reveals a novel conserved family that is homologous to bacterial and Aspergillus nucleobase transporters. Biochim Biophys Acta 1442: 353-360. Links Flament-Durand J, Brion J. 1985. Tanycytes: morphology and functions: a review. Int Rev Cytol 96: 121-155. Links García MA, Carrasco M, Godoy A, Reinicke K, Montecinos VP, Aguayo L, Tapia JC, Vera JC, Nualart F. 2001. Elevated expression of glucose transporter-1 in hypothalamic ependymal cells not involved in the formation of the brain-cerebrospinal fluid barrier. J Cell Biochem 80: 491-503. Links García MA, Millán C, Balmaceda-Aguilera C, Castro T, Montecinos H, Reinicke K, Vera JC, Oñate S, Nualart F. 2003. Hypothalamic ependymal-glial cells express the glucose transporter GLUT2, a protein involved in glucose sensing. J Neurochem 86: 709-724. Links Halliwell B. 1992. Reactive oxygen species and the central nervous system. J Neurochem 59: 1609-1623. Links Hediger MA. 2002. New view at C. Nat Med 8: 514-516. Links Hökfelt T, Foster G, Schultzberg M, Meister B, Schalling M, Goldstein M, Hemmings HC Jr, Ouimet C, Greengard P. 1988. DARPP-32 as a marker for D-1 dopaminoceptive cells in the rat brain: prenatal development and presence in glial elements (tanycytes) in the basal hypothalamus. Adv Exp Med Biol 235: 65-82. Links Karanth S, Yu W, Walczewska A, Mastronardi C, McCann S. 2000. Ascorbic acid acts as an inhibitory transmitter in the hypothalamus to inhibit stimulated luteinizing hormone-releasing hormone release by scavenging nitric oxide. Proc Natl Acad Sci USA 97: 1891-1899. Links

Korkok J, Yan R, Siushansian R, Dixon SJ, Wilson JX. 2000. Sodium-ascorbate cotransport controls intracellular ascorbate concentration in primary astrocyte cultures expressing the SVCT2 transporter. Brain Res 881: 144-151. Links Lutsenko E, Carcamo J, Golde DW. 2004. A human sodiumdependent vitamin C transporter 2 isoform acts as a dominantnegative inhibitor of ascorbic acid transport. Mol Cell Biol 24: 3150-3156. Links Makar TK, Nedergaard M, Preuss A, Gelbard AS, Perumal AS, Cooper AJ. 1994. Vitamin E, ascorbate, glutathione, glutathione disulfide, and enzymes of glutathione metabolism in cultures of chick astrocytes and neurons: evidence that astrocytes play an important role in antioxidant processes in the brain. J Neurochem 62: 45-53. Links Malo C, Wilson JX. 2000. Glucose modulates vitamin C transport in adult human small intestinal brush border membrane vesicles. J Nutrition 130: 63-69. Links Maulen NP, Henríquez E, Kempe S, Cárcamo JG, Schmid-Kotsas A, Bachem M, Grünert A, Bustamante ME, Nualart F, Vera JC. 2003. Up-regulation and polarized expression of the sodiumascorbic acid transporter SVCT1 in post-confluent differentiated CaCo-2 cells. J Biol Chem 278: 9035-9041. Links Nualart F, Rodríguez EM. 1996. Immunohistochemical analysis of subcommissural organ-Reissner's fiber complex using antibodies against alkylated and deglycosylated of the bovine Reissner's fiber. Cell Tissue Res 286: 23-31. Links Nualart F, Hein S, Rodríguez EM, Oksche A. 1991. Identification and partial characterization of the secretory glycoproteins of bovine subcommissural organ-Reissner's fiber complex. Evidence for the existence of two precursor forms. Mol Brain Res 11: 227238. Links Nualart F, Godoy K, Reinicke. 1999. Expression of the hexose transporters GLUT1 and GLUT2 during the early development of the human brain. Brain Res 824: 97-104. Links Nualart F, Rivas CI, Zhang RH, Guaiquil GH, Vera JC, Golde DW. 2003. Recycling of vitamin C by a bystander effect. J Biol Chem 278: 10128-10133. Links Oke AF, May L, Adams R. 1987. Ascorbic acid distribution patterns in human brain. Ann NY Acad Sci 498: 1-12. Links Patterson JW, Mastin DW. 1951. Some effects of dehydroascorbic acid on the central nervous system. Am J Physiol 167: 119-126. Links Peruzzo B, Pastor F, Blázquez J, Schöbitz K, Peláez B, Amat P, Rodríguez E. 2000. A second look at the barriers of the medial basal hypothalamus. Exp Brain Res 132: 10-26. Links Plummer TH Jr, Elder JH, Alexander S, Phelan AW, Tarentino AL. 1984. Demonstration of peptide: N-glycosidase F activity in endo-beta-N-acetylglucosaminidase F preparations. J Biol Chem 259: 10700-10704. Links Qutobet S, Dixon SJ, Wilson JX. 1998. Insulin stimulates vitamin C recycling and ascorbate accumulation in osteoblastic cells. Endocrinology 139: 51-56. Links Rajan DP, Huang W, Dutta B, Devoe LD, Leibach FH, Ganapathy V, Prasad PD. 1999. Human placental sodium-dependent vitamin C transporter (SVCT2): molecular cloning and transport function. Biochem Biophys Res Commun 262: 762-768. Links Rettori V, Belova N, Dees WL, Nyberg CL, Gimeno M, McCann SM. 1993. Role of nitric oxide in the control of luteinizing hormone-releasing hormone release in vivo and in vitro. Proc Natl Acad Sci USA 90: 10130-10134. Links

Rice ME. 2000. Ascorbate regulation and its neuroprotective role in the brain. Trends Neurosci 23: 209-216. Links Rodríguez EM, González CB, Dennoy L. 1979. Cellular organization of the lateral and postinfundibular regions of the median eminence in the rat. Cell Tissue Res 201: 377-408. Links Rose RC. 1988. Transport of ascorbic acid and other watersoluble vitamins. Biochim Biophys Acta 947: 335-366. Links Rose RC, Choi JL, Bode AM. 1992. Short term effects of oxidized ascorbic acid on bovine corneal endothelium and human placenta. Life Sci 50: 1543-1549. Links Rumsey SC, Kwon O, Xu GW, Burant CF, Simpson I, Levine M. 1997. Glucose transporter isoforms GLUT1 and GLUT3 transport dehydroascorbic acid. J Biol Chem 272: 18982-18989. Links Schenko JO, Miller E, Gaddis R Adams RN. 1982. Homeostatic control of ascorbate concentration in CNS extracellular fluid. Brain Res 253: 353-356. Links Schreiber M, Trojan S. 1991. Ascorbic acid in the brain. Physiol Res 40: 413-418. Links Siliprandi L, Vanni P, Kessler M, Semenza G. 1979. Na+dependent, electroneutral L-ascorbate transport across brush border membrane vesicles from guinea pig small intestine. Biochim Biophys Acta 552: 129-142. Links Siow RC, Sato H, Leake DS, Pearson JD, Bannai S, Mann GE. 1998. Vitamin C protects human arterial smooth muscle cells against atherogenic lipoproteins: effects of antioxidant vitamins C and E on oxidized LDL-induced adaptive increases in cystine transport and glutathione. Arterioscler Thromb Vasc Biol 18: 1662-1670. Links Siushansian R, Tao L, Dixon SJ, Wilson JX. 1997. Cerebral astrocytes transport ascorbic acid and dehydroascorbic acid through distinct mechanism regulated by cyclic AMP. J Neurochem 68: 2378-2385. Links Spector R, Lorenzo AV. 1973. Ascorbic acid homeostasis in the central nervous system. Am J Physiol 225: 757-763. Links Spielholz C, Golde DW, Houghton AN, Nualart F, Vera JC. 1997. Increased facilitated transport of dehydroascorbic acid without changes in sodium-dependent ascorbate transport in human melanoma cells. Cancer Res 57: 2529-2537. Links Tsukaguchi H, Tokui T, Mackenzie B, Berger UV, Chen X, Wang Y, Brubaker RF, Hediger MA. 1999. A family of mammalian Na+-dependent L-ascorbic acid transporters. Nature 399: 70-75. Links Vera JC, Rivas CI, Fischbarg J, Golde D. 1993. Mammalian facilitative hexose transporters mediate the transport of dehydroascorbic acid. Nature 364: 79-82. Links Wang H, Dutta B, Huang W, Devoe LD, Leibach FH, Ganapathy V, Prasad PD. 1999. Human Na+-dependent vitamin C transporter 1 (hSVCT1): primary structure, functional characteristics and evidence for a non-functional splice variant. Biochim Biophys Acta 1461: 1-9. Links Wang Y, Mackenzie B, Tsukaguchi H, Weremowicz S, Morton CC, Hediger MA. 2000. Human vitamin C (L-ascorbic acid) transporter SVCT1. Biochem Biophys Res Commun 267: 488494. Links Welch RW, Bergsten P, Butler JD, Levine M. 1993. Ascorbic acid accumulation and transport in human fibroblasts. Biochem J 294: 505-510. Links Welch RW, Wang Y, Crossman A Jr, Park JB, Kirk KL, Levine M. 1995. Accumulation of vitamin C (ascorbate) and its oxidized metabolite dehydroascorbic acid occurs by separate mechanism. J Biol Chem 270: 12584-12592. Links

Wilson JX. 1987. Antioxidant defense of the brain: a role for astrocytes. Can J Physiol Pharm 75: 1149-1163. Links Wilson JX. 1989. Ascorbic acid uptake by high-affinity sodiumdependent mechanism in cultured rat astrocytes. J Neurochem 53: 1064-1071. Links Wilson JX, Peters CE, Sitar SM, Daoust P, Gelb AW. 2000. Glutamate stimulates ascorbate transport by astrocytes. Brain Res 858: 61-66. Links Witenberg B, Kletter Y, Kalir HH, Raviv Z, Fening E, Nagler A, Halperin D, Fabian I. 1999. Ascorbic acid inhibits apoptosis induced by × irradiation in HL-60 myeloid leukemia cells. Radiat Res 152: 468-478. Links Yan J, Studer L, McKay RD. 2001. Ascorbic acid increases the yield of dopaminergic neurons derived from basic fibroblast growth factor expanded mesencephalic precursors. J Neurochem 76: 307-311. Links Zreik TG, Kodaman PH, Jones EE, Olive DL, Behrman H. 1999. Identification and characterization of an ascorbic acid transporter in human granulosa-lutein cells. Mol Hum Reprod 5: 299-302. Links